Retroviral vectors are capable of integrating transgenes into host chromosomes. However, the safety of retroviral vectors is a major issue since they integrate into the human genome at random locations and thus can potentially activate proto-oncogenes, triggering the development of cancer [1]. Thus, to aid the emergence of gene therapy it is necessary to develop safe strategies that allow therapeutic genes to be inserted into predetermined sites in the cellular genome.

The adeno-associated virus (AAV) genome is a 4.8 kb (kilobase) linear, single-stranded DNA, which contains a 145 nucleotide (nt) inverted terminal repeat (ITR) sequence at both ends (Figure 1A). The AAV preferentially integrates into the AAVS1 site on chromosome 19 (19q13.4). The AAVRep78 or Rep68 required for AAVS1-specific integration. [2] Rep78/68 specifically binds to a sequence motif within the ITR, the Rep binding site (RBS), and cleaves it in a site- and strand-specific manner at the terminal resolution site (trs) located 16 nt upstream of the RBS. A sequence homologous to the RBS/trs is present in the AAVS1 locus (Figure 1). The AAVS1 RBS motif is located 14 nt upstream of the translation initiation codon for the protein phosphatase 1 regulatory inhibitor subunit 12C gene (PPP1R12C), which is also called MBS85(myosin binding subunit 85). MBS85 is essential for actin depolymerization and cell motility. Site-specific integration using the AAV machinery enables the targeting of a therapeutic gene to a predetermined site, thereby minimizing the risk caused by insertional mutagenesis. Preferential integration, using an AAV plasmid-based system, has been demonstrated in cell lines derived from human embryonic kidney cells, HEK293 cells, [2] a human erythroid leukemia cell line, K562, [3] a human cervical carcinoma cell line [4] and human embryonic stem cells [5] (see Urabe et al., 2007 [6] for review).Rep-mediated integration system can efficiently insert ~200 kb foreign DNA into AAVS [7].

Marrow stromal cells are the cellular constituent of the marrow microenvironment and are essential for supporting hematopoiesis. In addition they are easy to obtain and rapidly expand in vitro. They are one of ideal targets for site-specific insertion of transgene since they can be vehicles to deliver therapeutic proteins.[8]

AIMS: We performed the AAVS1-targeted integration of a gene of interest into a bone marrow-derived stromal cell line, KM-102, as a model and examined the impact of the AAVS1-targeted insertion of the transgene on the host cells.

Plasmid construction: pRVK contains a 3.4kb AAVS1 sequence. pSVR68 expressing Rep68 under the control of the SV40 promoter (figure 2), was derived from pCMVR68. [2] The SV40 promoter was PCR amplified from pCMV-Tag2B (Stratagene, Pallo Alto, CA, USA) and inserted into pCMVR68 upstream of the Rep68 gene. pEB2, which harbors the CMV promoter, IRES (internal ribosome entry site), and a blasticidin S resistance gene (bsr) was described previously. [9] A humanized green fluorescent protein (hrGFP) gene was excised from phrGFPII-1 (Stratagene) and inserted into pEB2 upstream of the IRES. The hrGFP/bsr cassette was then inserted between the ITR sequences (pWGFPEB).

Cell culture and transfection: The KM-102 stromal cells [10] were maintained in McCoy's 5A medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% heat-inactivated fetal bovine serum (Sigma, St. Louis, MO, USA). The transfection of the KM-102 cells was performed with Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. Briefly a mixture of 2 µg of pSVR68 and an equal quantity of pWGFPEB was transfected into 3.5x105 KM-102 cells per well in a six-well plate. One day after transfection the cells were reseeded at an appropriate ratio. KM-102 cell clones were collected after being cultured for 14 days in the presence of 5 µg/ml of blasticidin S.

Southern blot analysis: Southern blot analysis was performed as described previously. [11] A 3.0-kb AccI fragment derived from pRVK was used as an AAVS1-specific probe (figure 1B), and a 3.0-kb NotI fragment from pWGFPEB was employed as a transgene probe (figure 2).

Quantitative real-time PCR for MBS85 mRNA: Total RNA was extracted using the RNeasy kit (Qiagen, Hilden, Germany). One microgram of total RNA was subjected to reverse transcription (SuperScript One-Step RT-PCR, Invitrogen) and qPCR (QuantiTect SYBR Green PCR Kit, Qiagen) in triplicate with primers for MBS85 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (table 1).

Site-specific integration in KM-102 cells: We analyzed 30 KM-102 clones by Southern blotting after digestion with EcoRV. Figure 3A shows a representative blot. The left panel is a blot that was hybridized with the AAVS1 probe and shows a 6.5-kb basal band in each lane (arrow). In some lanes, additional bands with slower mobility were detected, indicating the rearrangement of AAVS1. The right panel shows the same blot re-probed with a transgene-specific probe. Bands that hybridized with both the AAVS1 and transgene probes are indicated by arrowheads. Clones K4, K24, and K25 were considered to harbor the transgene at the AAVS1 locus and also confirmed to express GFP (figure 4A).

To verify that the GFP transgene had been inserted into the AAVS1 locus in these clones, we tried to PCR-amplify the junction sequence with primers specific to the CMV promoter or the bsr gene and an AAVS1-specific primer (figures 1B, 2, Table 1). In the K4 cells, the GFP/bsr transgene cassette was inserted upstream of the initiation codon for MBS85 in reverse orientation (figures 1 and 3B). In the K24 cells, the transgene cassette was inserted in the same direction as the MBS85 gene and it appeared that the first exon had been deleted. The breakpoints detected in the K4 and K24 cells are consistent with those described in previous reports. [4] [12] [13] No sequence was amplified from the K25 cells.

Impact of AAVS1 integration on the expression of MBS85: Since the AAVS1 site overlaps with the MBS85 gene, the insertion of foreign DNA into AAVS1 disrupts the MBS85 gene. Integration at AAVS1 also results in the deletion of the AAVS1 locus. [4] [12] [13] We quantified the expression of MBS85 mRNA relative to GAPDH mRNA using real-time RT-qPCR. The clones that harbored the transgene at the AAVS1 site (K4, K24, K25 cells) showed decreased MBS85 mRNA levels compared to the parental KM-102 cells and the K23 cells, which contained the transgene at a non-AAVS1 site (figure 4B).

G-CSF levels in the culture supernatants: KM-102 cells produce G-CSF upon stimulation with IL-1. [14] To examine the phenotypic changes induced in KM-102 cells after their transfection with a GFP plasmid and clonal expansion, we investigated the responsiveness of the clones to IL-1. Parental KM-102 cells and the transgenic clones were cultured in the presence of IL-1( (10 ng/ml), and then the G-CSF levels in the culture supernatants were measured. The clones secreted G-CSF after exposure to IL-1ß like the KM-102 cells (figure 4C). This result suggested that the incorporation of foreign DNA into the AAVS1 site did not affect the ability of the cells to respond to IL-1ß.

In the present study, we tested AAVS1-targeted insertion of transgene in KM-102 cells, a cell line derived from marrow stromal cells, and were able to specifically insert a GFP gene into the AAVS1 locus. Southern blot analysis demonstrated the AAVS1-specific integration of the transgene into KM-102 cells in three out of thirty clones (10%), which was confirmed by PCR analysis of the junction sequences between the transgene plasmid and the AAVS1 sequences in two clones. We failed to amplify the junction in K25 cells, which may be due to more extensive deletion of the MBS85 genome in K25 cells beyond the AAVS1 primer annealing site (figure 3A). Recently integration of AAV into AAVS1 has been reported to accompany partial duplication of the MBS85 gene and can preserve the normal expression of its mRNA. [15] They employed wild-type AAV and an AAV vector expressing GFP for targeted insertion in HeLa cells and mouse ES cells, whereas we used plasmid DNA in KM-102 cells. The different AAVS1 targeting strategy may affect the restoration of the MBS85 genome. In addition the same group had reported previously that the integration of AAV caused extensive deletion over the adjacent gene TNNT1. [16] Thorough understanding of Rep-mediated integration will be required for safe gene insertion technology based on AAV.

The insertion of foreign DNA into the AAVS1 site inevitably edits one allele of the MBS85 gene. However, KM-102 cells containing the transgene at AAVS1 did not appear to show phenotypic changes although their MBS85 mRNA level decreased (figure 4B). Therefore, the disruption of one of the MBS85 gene's alleles might not have serious effects in KM-102 cells although the impact of such MBS85 gene disruption should be thoroughly investigated. The integration of a transgene into a predetermined site in the human genome, which minimizes the risk caused by insertion mutagenesis, is particularly important for the application of gene transfer to proliferating cells such as stem cells. Our strategy for delivering a therapeutic gene into the AAVS1 locus using the AAV components ITR and Rep is an attractive approach for gene therapy and regenerative medicine.

The AAVS1 site is a safe harbor for transgene insertion although it results in impaired MBS85 expression under some condition.

1. Hacein-Bey-Abina S, Von Kalle C, Schmidt M, McCormack MP, Wulffraat N, Leboulch P, et al. LMO2-associated clonal T cell proliferation in two patients after gene therapy for SCID-X1. Science 2003;302(5644):415-419.   [CrossRef]   [Pubmed]    
2. Surosky RT, Urabe M, Godwin SG, McQuiston SA, Kurtzman GJ, Ozawa K, et al. Adeno-associated virus Rep proteins target DNA sequences to a unique locus in the human genome. J Virol 1997;71(10):7951-7959.   [CrossRef]   [Pubmed]    
3. Kogure K, Urabe M, Mizukami H, Kume A, Sato Y, Monahan J, et al. Targeted integration of foreign DNA into a defined locus on chromosome 19 in K562 cells using AAV-derived components. Int J Hematol 2001;73(4):469-475.   [CrossRef]   [Pubmed]    
4. Tsunoda H, Hayakawa T, Sakuragawa N, Koyama H. Site-specific integration of adeno-associated virus-based plasmid vectors in lipofected HeLa cells. Virology 2000;268(2):391-401.   [CrossRef]   [Pubmed]    
5. Smith JR, Maguire S, Davis LA, Alexander M, Yang F, Chandran S, et al. Robust, persistent transgene expression in human embryonic stem cells is achieved with AAVS1-targeted integration. Stem Cells 2008;26(2):496-504.   [CrossRef]   [Pubmed]    
6. Urabe M, Obara, Y., Ito, T., Mizukami, H., Kume, A., Ozawa, K. Targeted insertion of transgene into a specific site on chromosome 19 by using adeno-associated virus integration machinery.vol. 3. In: Bertolotti R, Ozawa, K, editors. Progress in Gene Therapy. Singapore: World Scientific; 2007. p.19-46.   [CrossRef]   [Pubmed]    
7. Howden SE, Voullaire L, Wardan H, Williamson R, Vadolas J. Site-specific, Rep-mediated integration of the intact beta-globin locus in the human erythroleukaemic cell line K562. Gene Ther 2008;15(20):1372-1383.   [CrossRef]   [Pubmed]    
8. Hurwitz DR, Kirchgesser M, Merrill W, Galanopoulos T, McGrath CA, Emami S, et al. Systemic delivery of human growth hormone or human factor IX in dogs by reintroduced genetically modified autologous bone marrow stromal cells. Hum Gene Ther 1997;8(2):137-156.   [CrossRef]   [Pubmed]    
9. Urabe M, Hasumi Y, Ogasawara Y, Matsushita T, Kamoshita N, Nomoto A, et al. A novel dicistronic AAV vector using a short IRES segment derived from hepatitis C virus genome. Gene 1997;200(1-2):157-162.   [CrossRef]   [Pubmed]    
10. Harigaya K, Handa H. Generation of functional clonal cell lines from human bone marrow stroma. Proc Natl Acad Sci USA 1985;82(10):3477-3480.   [CrossRef]   [Pubmed]    
11. Urabe M, Kogure K, Kume A, Sato Y, Tobita K, Ozawa K. Positive and negative effects of adeno-associated virus Rep on AAVS1-targeted integration. J Gen Virol 2003;84(Pt 8):2127-2132.   [CrossRef]   [Pubmed]    
12. Pieroni L, Fipaldini C, Monciotti A, Cimini D, Sgura A, Fattori E, et al. Targeted integration of adeno-associated virus-derived plasmids in transfected human cells. Virology 1998;249(2):249-259.   [CrossRef]   [Pubmed]    
13. Samulski RJ, Zhu X, Xiao X, Brook JD, Housman DE, Epstein N, et al. Targeted integration of adeno-associated virus (AAV) into human chromosome 19. Embo J 1991;10(12):3941-3950.   [CrossRef]   [Pubmed]    
14. Watari K, Ozawa K, Tajika K, Tojo A, Tani K, Kamachi S, et al. Production of human granulocyte colony stimulating factor by various kinds of stromal cells in vitro detected by enzyme immunoassay and in situ hybridization. Stem Cells 1994;12(4):416-423.   [CrossRef]   [Pubmed]    
15. Henckaerts E, Dutheil N, Zeltner N, Kattman S, Kohlbrenner E, Ward P, et al. Site-specific integration of adeno-associated virus involves partial duplication of the target locus. Proc Natl Acad Sci USA 2009;106(18):7571-7576.   [CrossRef]   [Pubmed]    
16. Dutheil N, Shi F, Dupressoir T, Linden RM. Adeno-associated virus site-specifically integrates into a muscle-specific DNA region. Proc Natl Acad Sci USA 2000;97(9):4862-4866.   [CrossRef]   [Pubmed]